other-emerald
other-emerald16mo ago

0,35 per minute?

Why is the cost per minute so high?
No description
11 Replies
Ask Inkeep
Ask Inkeep16mo ago
Hi @wielki asasyn, I'm an AI assistant for Vapi. I'll analyze available sources and get back to you shortly. ----------------------------- While I generate an answer, here are some sources that might be helpful: * Billing Examples * Estimating Costs * Pricing Overview * Cost Routing -----------------------------
Ask Inkeep
Ask Inkeep16mo ago
Answer The cost per minute can vary significantly based on several factors in your voice pipeline, such as the choice of providers for transcription, text-to-speech, and the language model, as well as the actual usage like the number of tokens or characters processed. Here’s a breakdown that might explain higher costs: 1. Transcription and Voice Costs: Different providers have different rates for transcription and voice synthesis. For instance, providers like Deepgram for transcription and PlayHT for voice synthesis charge per minute of usage. If your application requires high-quality voices or handles complex audio inputs, these costs can add up quickly (1) . 2. Custom Model Usage: If you’re using a custom model, the costs can be significantly higher depending on the complexity and computational requirements of the model. The example in the documentation suggests a custom model could cost around $0.04/min, which contributes to the total cost (2) . 3. Telephony Costs: If your voice application involves phone calls, providers like Twilio have per-minute charges for handling these calls (1) . 4. Vapi Platform Fees: Vapi itself charges a standard rate of $0.05/min for managing the orchestration of these services (3) . For a detailed and accurate cost estimate based on your specific usage, you might want to check the Vapi dashboard or use the cost estimation tools provided in the Vapi documentation (1) . If you're experiencing unexpected high costs, consider reviewing the amount of data processed per call or adjusting the providers or services used to see if a more cost-effective setup can be achieved.
Ask Inkeep
Ask Inkeep16mo ago
Got it. Tagging @Vapi Team for additional help.
old-apricot
old-apricot16mo ago
Do you still need any help? @wielki asasyn
other-emerald
other-emeraldOP16mo ago
Yeah @Sahil can you explain the huge cost per minute? It used to be 0.10-15/min and now it's triple that?
old-apricot
old-apricot16mo ago
It is not the actual cost. It just an approximate like as the size of prompt increases price also increases as it gets appended to all the messages. We just improved our approximation model. That’s all
other-emerald
other-emeraldOP16mo ago
No description
other-emerald
other-emeraldOP16mo ago
No description
other-emerald
other-emeraldOP16mo ago
This is from two different assitants What's the reason that it varies this much?
vicious-gold
vicious-gold16mo ago
prompt size bruv
Ask Inkeep
Ask Inkeep16mo ago
Glad I could be helpful. Feel free to create a new thread with any new questions you may have.

Did you find this page helpful?